The loss of a spouse is one of the most severe stressors for individuals of any age. This study proposes to evaluate a specific theoretical model which hypothesizes that: 1) There is a direct effect of social support on the well-being of widows and widowers. 2) There is an indirect effect of social support on the well-being of widows and widowers through perceived stress. 3) There is an indirect effect of social support on the well-being of widows and widowers through coping behaviors. 4) There is an indirect effect of social support on the well-being of widows and widowers through health practices. Changes in the kind, amount, and quality of social support as well as fluctuations in perceived stress levels, coping behaviors, health practices and well-being across the bereavement period also will be identified and described. Using a structured interview, data will be collected from 250 middle-age and older widows and widowers at 6 weeks, 6 months, 12 months, 18 months, and 24 months following the death of a spouse. Path analysis will be used to evaluate the proposed theoretical model. Patterns of interrelationships among variables across time will be examined through a series of cross-lagged panel correlations. The relationship of selected demographic charcteristics to changes in the variables during the bereavement period will be explored.